Multi-omic data integration and analyses for biomarker discovery of spontaneous preterm birth phenotypes

BackgroundPreterm birth, delivery <37 weeks of gestation, is a global health concern affecting 1.3 million infants annually. A higher morbidity burden is associated with infants born <34 weeks of gestation. A more robust biomarker of spontaneous preterm birth needs to be identified. To our knowledge, this study uniquely integrated three ome-wide datasets prospectively collected from the same individuals within a single, defined UK cohort. The study aimed to integrate genomic, transcriptomic, and metabolomic data collected from the same cohort of women for biomarker discovery. MethodsPregnant women with a history of a previous spontaneous preterm birth (sPTB) <34 weeks were recruited in Liverpool in a subsequent pregnancy at 16 and/or 20 weeks of gestation. Pregnancy outcomes were followed up and categorised into the different clinical subgroups of SPTB <34 weeks, PPROM <34 weeks and term delivery >37 weeks (controls). Blood samples were profiled for genomics, transcriptomics and metabolomics. ANOVA analyses were performed at each gestational timepoint. Network enrichment analyses was performed on significant genes. ResultsAfter multi-omic data integration, 43 women at week 16 and 40 women at week 20 of gestation had all three genomics, transcriptomics and metabolomic profiling available. Multiple significant transcripts were detected (p<0.05) from the ANOVA analyses, though these were mostly in non-coding regions. ConclusionThree different omic data, genomics, transcriptomics and metabolomics, were integrated across the same individuals for biomarker analyses. Molecular signatures were detected that could lead to understanding the pathways involved in the different subgroups of preterm birth: SPTB and PPROM.